[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] Oops when modprobing ivtv outside of dom0



> On 7/4/05, Keir Fraser <Keir.Fraser@xxxxxxxxxxxx> wrote:
> 
> > That's weird. It's crashing somewhere in ivtv_probe(), yet one of the
> > first things that function does is print a line of debug output. But
> > that output hasn;t appeared in your trace. :-/
> >
> > Maybe try adding an explicit printk at the start of ivtv_probe and see
> > if that appears:
> >   printk(KERN_ALERT "**********\n");
> 
> You mean the "Found card #" message? That did show up on the one that
> I modprobe'd with ivtv_debug=255.
> 
> Here's the dmesg output with the printk added, and with ivtv_debug=255:

Ah, I see the bug. It's in both 2.0 and 3.3 (the unstable version) so
it's worth pointing out to the developers. The following code in
ivtv_setup_pci() is broken:

        if (pci_bus != NULL && pci_bus->vendor == PCI_VENDOR_ID_VIA) {
                ....
        } else
                IVTV_KERN_INFO("XXX PCI device: 0x%04x vendor: 0x%04x\n",
                                pci_bus->device, pci_bus->vendor);

The else clause should be 'else if (pci_bus != NULL)'. Otherwise the
debug IVTV_KERN_INFO line crashes trying to dereference a NULL
pointer. :-)

Actually, I think maybe they could get rid of that entire block of
code -- nothing seems to depend on via_fix, which is the variable that
it sets up.

 -- Keir

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.